mlib_v_ImageConstLogic.h revision 0
0N/A/*
0N/A * Copyright 2003 Sun Microsystems, Inc. All Rights Reserved.
0N/A * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
0N/A *
0N/A * This code is free software; you can redistribute it and/or modify it
2362N/A * under the terms of the GNU General Public License version 2 only, as
0N/A * published by the Free Software Foundation. Sun designates this
2362N/A * particular file as subject to the "Classpath" exception as provided
0N/A * by Sun in the LICENSE file that accompanied this code.
0N/A *
0N/A * This code is distributed in the hope that it will be useful, but WITHOUT
0N/A * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
0N/A * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
0N/A * version 2 for more details (a copy is included in the LICENSE file that
0N/A * accompanied this code).
0N/A *
0N/A * You should have received a copy of the GNU General Public License version
0N/A * 2 along with this work; if not, write to the Free Software Foundation,
0N/A * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
2362N/A *
2362N/A * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
2362N/A * CA 95054 USA or visit www.sun.com if you need additional information or
0N/A * have any questions.
0N/A */
0N/A
1693N/A#ifndef __MLIB_V_IMAGECONSTLOGIC_H
0N/A#define __MLIB_V_IMAGECONSTLOGIC_H
0N/A
3171N/A
0N/A#if defined ( VIS )
0N/A#if VIS >= 0x200
0N/A#error This include file can be used with VIS 1.0 only
0N/A#endif /* VIS >= 0x200 */
0N/A#endif /* defined ( VIS ) */
0N/A
0N/A#include <mlib_image.h>
0N/A#include <vis_proto.h>
0N/A#include <mlib_ImageCheck.h>
0N/A#include <mlib_ImageLogic_proto.h>
0N/A#include <mlib_v_ImageLogic_proto.h>
0N/A
0N/A#ifdef __cplusplus
0N/Aextern "C" {
0N/A#endif /* __cplusplus */
0N/A
0N/A/*
0N/A * Macro definitions for VIS version image logical functions.
0N/A */
1693N/A
1693N/A/***************************************************************/
0N/A
0N/A#define VALIDATE() \
0N/A mlib_u8 *sp, *sl; /* pointers for pixel and line of source */ \
0N/A mlib_u8 *dp, *dl;/* pointers for pixel and line of dst */ \
0N/A mlib_s32 width, height, type, nchannels; \
1693N/A mlib_s32 stride; /* for src */ \
1693N/A mlib_s32 strided; /* for dst */ \
0N/A mlib_u32 c01, c02, c03, c04; \
1693N/A mlib_d64 dc01, dc02, dc03; \
1693N/A \
0N/A MLIB_IMAGE_SIZE_EQUAL(dst,src); \
0N/A MLIB_IMAGE_TYPE_EQUAL(dst,src); \
0N/A MLIB_IMAGE_CHAN_EQUAL(dst,src); \
0N/A \
0N/A dp = (mlib_u8 *) mlib_ImageGetData(dst); \
0N/A sp = (mlib_u8 *) mlib_ImageGetData(src); \
0N/A height = mlib_ImageGetHeight(dst); \
1693N/A width = mlib_ImageGetWidth(dst); \
1693N/A stride = mlib_ImageGetStride(src); \
1693N/A strided = mlib_ImageGetStride(dst); \
1693N/A nchannels = mlib_ImageGetChannels(dst); \
0N/A type = mlib_ImageGetType(dst); \
0N/A \
0N/A if (type == MLIB_SHORT) { \
3171N/A width *= (2 * nchannels); \
0N/A if (nchannels == 1) { \
3171N/A c01 = c[0] & 0xFFFF; c01 |= (c01 << 16); \
0N/A dc01 = vis_to_double_dup(c01); \
0N/A } else if (nchannels == 2) { \
0N/A c01 = ((c[0] & 0xFFFF) << 16) | (c[1] & 0xFFFF); \
0N/A dc01 = vis_to_double_dup(c01); \
0N/A } else if (nchannels == 3) { \
0N/A c01 = ((c[0] & 0xFFFF) << 16) | (c[1] & 0xFFFF); \
0N/A c02 = ((c[2] & 0xFFFF) << 16) | (c01 >> 16); \
1693N/A c03 = (c01 << 16) | (c02 >> 16); \
5980N/A dc01= vis_to_double(c01, c02); \
0N/A dc02= vis_to_double(c03, c01); \
0N/A dc03= vis_to_double(c02, c03); \
1693N/A } else { \
1693N/A c01 = ((c[0] & 0xFFFF) << 16) | (c[1] & 0xFFFF); \
0N/A c02 = ((c[2] & 0xFFFF) << 16) | (c[3] & 0xFFFF); \
0N/A dc01= vis_to_double(c01, c02); \
0N/A } \
0N/A \
3171N/A } else if (type == MLIB_BYTE) { \
0N/A width *= nchannels; \
0N/A if (nchannels == 1) { \
0N/A c01 = c[0] & 0xFF; c01 |= (c01 << 8); \
0N/A c01 |= (c01 << 16); \
0N/A dc01 = vis_to_double_dup(c01); \
0N/A } else if (nchannels == 2) { \
1693N/A c01 = ((c[0] & 0xFF) << 8) | (c[1] & 0xFF); \
3171N/A c01 |= (c01 << 16); \
0N/A dc01 = vis_to_double_dup(c01); \
0N/A } else if (nchannels == 3) { \
0N/A c01 = ((c[0] & 0xFF) << 16) | ((c[1] & 0xFF) << 8) | \
0N/A (c[2] & 0xFF); \
0N/A c02 = (c01 << 16) | (c01 >> 8); \
0N/A c03 = (c01 << 24) | c01; \
0N/A c01 = (c01 << 8) | (c01 >> 16); \
0N/A dc01= vis_to_double(c01, c02); \
1693N/A dc02= vis_to_double(c03, c01); \
1693N/A dc03= vis_to_double(c02, c03); \
1693N/A } else { \
1693N/A c01 = ((c[0] & 0xFF) << 24) | ((c[1] & 0xFF) << 16) | \
1693N/A ((c[2] & 0xFF) << 8) | (c[3] & 0xFF); \
1693N/A dc01 = vis_to_double_dup(c01); \
1693N/A } \
1693N/A } else { \
1693N/A width *= (4 * nchannels); \
1693N/A if (nchannels == 1) { \
1693N/A c01 = c[0] & 0xFFFFFFFF; \
1693N/A dc01 = vis_to_double_dup(c01); \
1693N/A } else if (nchannels == 2) { \
1693N/A c01 = c[0] & 0xFFFFFFFF; c02 = c[1] & 0xFFFFFFFF; \
0N/A dc01 = vis_to_double(c01, c02); \
0N/A } else if (nchannels == 3) { \
0N/A c01 = c[0] & 0xFFFFFFFF; c02 = c[1] & 0xFFFFFFFF; \
1693N/A c03 = c[2] & 0xFFFFFFFF; \
1693N/A dc01= vis_to_double(c01, c02); \
1693N/A dc02= vis_to_double(c03, c01); \
0N/A dc03= vis_to_double(c02, c03); \
0N/A } else { \
0N/A c01 = c[0] & 0xFFFFFFFF; c02 = c[1] & 0xFFFFFFFF; \
0N/A c03 = c[2] & 0xFFFFFFFF; c04 = c[3] & 0xFFFFFFFF; \
0N/A dc01= vis_to_double(c01, c02); \
0N/A dc02= vis_to_double(c03, c04); \
1693N/A } \
1693N/A } \
1693N/A \
0N/A if ((width > stride) || (width > strided)) \
0N/A return MLIB_FAILURE
0N/A
0N/A/***************************************************************/
0N/A
0N/Astatic mlib_status mlib_v_ImageConstLogic(mlib_image *dst,
0N/A mlib_image *src,
0N/A mlib_s32 *c)
1693N/A{
1693N/A mlib_s32 i, j;
1693N/A mlib_s32 offdst, offsrc, emask;
0N/A mlib_d64 *dpp, *spp;
0N/A mlib_d64 sa1, sa2, da, sa;
1693N/A mlib_d64 ssa, ssa1, ssa2, sa3, sa4;
0N/A mlib_s32 amount;
1693N/A mlib_u8 *dend;
mlib_d64 c1, c2, c3;
VALIDATE();
if (nchannels == 3) {
if ((width == stride) && (width == strided) && ((width - (width / 3) * 3) == 0)) {
amount = height * width;
dend = dp + amount - 1;
offdst = ((mlib_addr) dp) & 7;
offsrc = ((mlib_addr) sp) & 7;
if (offsrc == offdst) {
/* prepare the destination addresses */
dpp = (mlib_d64 *) vis_alignaddr(dp, 0);
i = (mlib_u8 *) dpp - dp;
if (i != 0) {
vis_alignaddr((void *)(8 - offdst), 0);
c3 = vis_faligndata(dc03, dc01);
c1 = vis_faligndata(dc01, dc02);
c2 = vis_faligndata(dc02, dc03);
}
else {
c1 = dc01;
c2 = dc02;
c3 = dc03;
}
/* prepare the destination addresses */
spp = (mlib_d64 *) vis_alignaddr(sp, 0);
if (i != 0) {
/* generate edge mask for the start point */
emask = vis_edge8(dp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c3, sa1);
vis_pst_8(da, dpp++, emask);
i += 8;
}
#pragma pipeloop(0)
for (; i < amount - 24; i += 24) {
dpp[0] = VIS_CONSTLOGIC(c1, spp[0]);
dpp[1] = VIS_CONSTLOGIC(c2, spp[1]);
dpp[2] = VIS_CONSTLOGIC(c3, spp[2]);
dpp += 3;
spp += 3;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c1, sa1);
vis_pst_8(da, dpp++, emask);
i += 8;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c2, sa1);
vis_pst_8(da, dpp++, emask);
i += 8;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c3, sa1);
vis_pst_8(da, dpp, emask);
}
}
else {
/* prepare the destination addresses */
dpp = (mlib_d64 *) vis_alignaddr(dp, 0);
i = (mlib_u8 *) dpp - dp;
if (i != 0) {
vis_alignaddr((void *)(8 - offdst), 0);
c3 = vis_faligndata(dc03, dc01);
c1 = vis_faligndata(dc01, dc02);
c2 = vis_faligndata(dc02, dc03);
}
else {
c1 = dc01;
c2 = dc02;
c3 = dc03;
}
/* prepare the destination addresses */
spp = (mlib_d64 *) vis_alignaddr(sp, i);
sa1 = spp[0];
if (i != 0) {
/* generate edge mask for the start point */
emask = vis_edge8(dp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c3, sa);
vis_pst_8(da, dpp++, emask);
sa1 = sa2;
i += 8;
spp++;
}
#pragma pipeloop(0)
for (; i < amount - 24; i += 24) {
sa2 = spp[1];
ssa = vis_faligndata(sa1, sa2);
dpp[0] = VIS_CONSTLOGIC(c1, ssa);
sa3 = spp[2];
ssa1 = vis_faligndata(sa2, sa3);
dpp[1] = VIS_CONSTLOGIC(c2, ssa1);
sa4 = spp[3];
ssa2 = vis_faligndata(sa3, sa4);
dpp[2] = VIS_CONSTLOGIC(c3, ssa2);
sa1 = sa4;
dpp += 3;
spp += 3;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c1, sa);
vis_pst_8(da, dpp++, emask);
sa1 = sa2;
i += 8;
spp++;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c2, sa);
vis_pst_8(da, dpp++, emask);
sa1 = sa2;
i += 8;
spp++;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c3, sa);
vis_pst_8(da, dpp++, emask);
}
}
}
else {
sl = sp;
dl = dp;
amount = width;
for (j = 0; j < height; j++) {
dend = dp + amount - 1;
offdst = ((mlib_addr) dp) & 7;
offsrc = ((mlib_addr) sp) & 7;
if (offsrc == offdst) {
/* prepare the destination addresses */
dpp = (mlib_d64 *) vis_alignaddr(dp, 0);
i = (mlib_u8 *) dpp - dp;
if (i != 0) {
vis_alignaddr((void *)(8 - offdst), 0);
c3 = vis_faligndata(dc03, dc01);
c1 = vis_faligndata(dc01, dc02);
c2 = vis_faligndata(dc02, dc03);
}
else {
c1 = dc01;
c2 = dc02;
c3 = dc03;
}
/* prepare the destination addresses */
spp = (mlib_d64 *) vis_alignaddr(sp, 0);
if (i != 0) {
/* generate edge mask for the start point */
emask = vis_edge8(dp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c3, sa1);
vis_pst_8(da, dpp++, emask);
i += 8;
}
#pragma pipeloop(0)
for (; i < amount - 24; i += 24) {
dpp[0] = VIS_CONSTLOGIC(c1, spp[0]);
dpp[1] = VIS_CONSTLOGIC(c2, spp[1]);
dpp[2] = VIS_CONSTLOGIC(c3, spp[2]);
dpp += 3;
spp += 3;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c1, sa1);
vis_pst_8(da, dpp++, emask);
i += 8;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c2, sa1);
vis_pst_8(da, dpp++, emask);
i += 8;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c3, sa1);
vis_pst_8(da, dpp, emask);
}
}
else {
/* prepare the destination addresses */
dpp = (mlib_d64 *) vis_alignaddr(dp, 0);
i = (mlib_u8 *) dpp - dp;
if (i != 0) {
vis_alignaddr((void *)(8 - offdst), 0);
c3 = vis_faligndata(dc03, dc01);
c1 = vis_faligndata(dc01, dc02);
c2 = vis_faligndata(dc02, dc03);
}
else {
c1 = dc01;
c2 = dc02;
c3 = dc03;
}
/* prepare the destination addresses */
spp = (mlib_d64 *) vis_alignaddr(sp, i);
sa1 = spp[0];
if (i != 0) {
/* generate edge mask for the start point */
emask = vis_edge8(dp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c3, sa);
vis_pst_8(da, dpp++, emask);
sa1 = sa2;
i += 8;
spp++;
}
#pragma pipeloop(0)
for (; i < amount - 24; i += 24) {
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
dpp[0] = VIS_CONSTLOGIC(c1, sa);
sa1 = spp[2];
sa = vis_faligndata(sa2, sa1);
dpp[1] = VIS_CONSTLOGIC(c2, sa);
sa2 = spp[3];
sa = vis_faligndata(sa1, sa2);
dpp[2] = VIS_CONSTLOGIC(c3, sa);
sa1 = sa2;
dpp += 3;
spp += 3;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c1, sa);
vis_pst_8(da, dpp++, emask);
sa1 = sa2;
i += 8;
spp++;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c2, sa);
vis_pst_8(da, dpp++, emask);
sa1 = sa2;
i += 8;
spp++;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c3, sa);
vis_pst_8(da, dpp++, emask);
}
}
sp = sl += stride;
dp = dl += strided;
}
}
}
else if ((type != MLIB_INT) || (nchannels != 4)) {
if ((width == stride) && (width == strided)) {
amount = height * width;
dend = dp + amount - 1;
offdst = ((mlib_addr) dp) & 7;
offsrc = ((mlib_addr) sp) & 7;
if (offsrc == offdst) {
/* prepare the destination addresses */
dpp = (mlib_d64 *) vis_alignaddr(dp, 0);
i = (mlib_u8 *) dpp - dp;
if (i != 0) {
vis_alignaddr((void *)(8 - offdst), 0);
c1 = vis_faligndata(dc01, dc01);
}
else {
c1 = dc01;
}
/* prepare the destination addresses */
spp = (mlib_d64 *) vis_alignaddr(sp, 0);
if (i != 0) {
/* generate edge mask for the start point */
emask = vis_edge8(dp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c1, sa1);
vis_pst_8(da, dpp++, emask);
i += 8;
}
#pragma pipeloop(0)
for (; i < amount - 8; i += 8) {
*dpp++ = VIS_CONSTLOGIC(c1, *spp);
spp++;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa1 = *spp;
da = VIS_CONSTLOGIC(c1, sa1);
vis_pst_8(da, dpp, emask);
}
}
else {
/* prepare the destination addresses */
dpp = (mlib_d64 *) vis_alignaddr(dp, 0);
i = (mlib_u8 *) dpp - dp;
if (i != 0) {
vis_alignaddr((void *)(8 - offdst), 0);
c1 = vis_faligndata(dc01, dc01);
}
else {
c1 = dc01;
}
/* prepare the destination addresses */
spp = (mlib_d64 *) vis_alignaddr(sp, i);
sa1 = spp[0];
if (i != 0) {
/* generate edge mask for the start point */
emask = vis_edge8(dp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c1, sa);
vis_pst_8(da, dpp++, emask);
sa1 = sa2;
i += 8;
spp++;
}
#pragma pipeloop(0)
for (; i < amount - 8; i += 8) {
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
*dpp++ = VIS_CONSTLOGIC(c1, sa);
sa1 = sa2;
spp++;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c1, sa);
vis_pst_8(da, dpp, emask);
}
}
}
else {
sl = sp;
dl = dp;
amount = width;
for (j = 0; j < height; j++) {
dend = dp + amount - 1;
offdst = ((mlib_addr) dp) & 7;
offsrc = ((mlib_addr) sp) & 7;
if (offsrc == offdst) {
/* prepare the destination addresses */
dpp = (mlib_d64 *) vis_alignaddr(dp, 0);
i = (mlib_u8 *) dpp - dp;
if (i != 0) {
vis_alignaddr((void *)(8 - offdst), 0);
c1 = vis_faligndata(dc01, dc01);
}
else {
c1 = dc01;
}
/* prepare the destination addresses */
spp = (mlib_d64 *) vis_alignaddr(sp, 0);
if (i != 0) {
/* generate edge mask for the start point */
emask = vis_edge8(dp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c1, sa1);
vis_pst_8(da, dpp++, emask);
i += 8;
}
#pragma pipeloop(0)
for (; i < amount - 8; i += 8) {
*dpp++ = VIS_CONSTLOGIC(c1, *spp);
spp++;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa1 = *spp;
da = VIS_CONSTLOGIC(c1, sa1);
vis_pst_8(da, dpp, emask);
}
}
else {
/* prepare the destination addresses */
dpp = (mlib_d64 *) vis_alignaddr(dp, 0);
i = (mlib_u8 *) dpp - dp;
if (i != 0) {
vis_alignaddr((void *)(8 - offdst), 0);
c1 = vis_faligndata(dc01, dc01);
}
else {
c1 = dc01;
}
/* prepare the destination addresses */
spp = (mlib_d64 *) vis_alignaddr(sp, i);
sa1 = spp[0];
if (i != 0) {
/* generate edge mask for the start point */
emask = vis_edge8(dp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c1, sa);
vis_pst_8(da, dpp++, emask);
sa1 = sa2;
i += 8;
spp++;
}
#pragma pipeloop(0)
for (; i < amount - 8; i += 8) {
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
*dpp++ = VIS_CONSTLOGIC(c1, sa);
sa1 = sa2;
spp++;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c1, sa);
vis_pst_8(da, dpp, emask);
}
}
sp = sl += stride;
dp = dl += strided;
}
}
}
else {
if ((width == stride) && (width == strided)) {
amount = height * width;
dend = dp + amount - 1;
offdst = ((mlib_addr) dp) & 7;
offsrc = ((mlib_addr) sp) & 7;
if (offsrc == offdst) {
/* prepare the destination addresses */
dpp = (mlib_d64 *) vis_alignaddr(dp, 0);
i = (mlib_u8 *) dpp - dp;
if (i != 0) {
vis_alignaddr((void *)(8 - offdst), 0);
c2 = vis_faligndata(dc02, dc01);
c1 = vis_faligndata(dc01, dc02);
}
else {
c1 = dc01;
c2 = dc02;
}
/* prepare the destination addresses */
spp = (mlib_d64 *) vis_alignaddr(sp, 0);
if (i != 0) {
/* generate edge mask for the start point */
emask = vis_edge8(dp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c2, sa1);
vis_pst_8(da, dpp++, emask);
i += 8;
}
#pragma pipeloop(0)
for (; i < amount - 16; i += 16) {
dpp[0] = VIS_CONSTLOGIC(c1, spp[0]);
dpp[1] = VIS_CONSTLOGIC(c2, spp[1]);
dpp += 2;
spp += 2;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c1, sa1);
vis_pst_8(da, dpp++, emask);
i += 8;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa1 = *spp;
da = VIS_CONSTLOGIC(c2, sa1);
vis_pst_8(da, dpp++, emask);
}
}
else {
/* prepare the destination addresses */
dpp = (mlib_d64 *) vis_alignaddr(dp, 0);
i = (mlib_u8 *) dpp - dp;
if (i != 0) {
vis_alignaddr((void *)(8 - offdst), 0);
c2 = vis_faligndata(dc02, dc01);
c1 = vis_faligndata(dc01, dc02);
}
else {
c1 = dc01;
c2 = dc02;
}
/* prepare the destination addresses */
spp = (mlib_d64 *) vis_alignaddr(sp, i);
sa1 = spp[0];
if (i != 0) {
/* generate edge mask for the start point */
emask = vis_edge8(dp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c2, sa);
vis_pst_8(da, dpp++, emask);
sa1 = sa2;
i += 8;
spp++;
}
#pragma pipeloop(0)
for (; i < amount - 16; i += 16) {
sa2 = spp[1];
ssa = vis_faligndata(sa1, sa2);
dpp[0] = VIS_CONSTLOGIC(c1, ssa);
sa3 = spp[2];
ssa1 = vis_faligndata(sa2, sa3);
dpp[1] = VIS_CONSTLOGIC(c2, ssa1);
sa1 = sa3;
dpp += 2;
spp += 2;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c1, sa);
vis_pst_8(da, dpp++, emask);
sa1 = sa2;
i += 8;
spp++;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c2, sa);
vis_pst_8(da, dpp++, emask);
}
}
}
else {
sl = sp;
dl = dp;
amount = width;
for (j = 0; j < height; j++) {
dend = dp + amount - 1;
offdst = ((mlib_addr) dp) & 7;
offsrc = ((mlib_addr) sp) & 7;
if (offsrc == offdst) {
/* prepare the destination addresses */
dpp = (mlib_d64 *) vis_alignaddr(dp, 0);
i = (mlib_u8 *) dpp - dp;
if (i != 0) {
vis_alignaddr((void *)(8 - offdst), 0);
c2 = vis_faligndata(dc02, dc01);
c1 = vis_faligndata(dc01, dc02);
}
else {
c1 = dc01;
c2 = dc02;
}
/* prepare the destination addresses */
spp = (mlib_d64 *) vis_alignaddr(sp, 0);
if (i != 0) {
/* generate edge mask for the start point */
emask = vis_edge8(dp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c2, sa1);
vis_pst_8(da, dpp++, emask);
i += 8;
}
#pragma pipeloop(0)
for (; i < amount - 16; i += 16) {
dpp[0] = VIS_CONSTLOGIC(c1, spp[0]);
dpp[1] = VIS_CONSTLOGIC(c2, spp[1]);
dpp += 2;
spp += 2;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa1 = *spp++;
da = VIS_CONSTLOGIC(c1, sa1);
vis_pst_8(da, dpp++, emask);
i += 8;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa1 = *spp;
da = VIS_CONSTLOGIC(c2, sa1);
vis_pst_8(da, dpp++, emask);
}
}
else {
/* prepare the destination addresses */
dpp = (mlib_d64 *) vis_alignaddr(dp, 0);
i = (mlib_u8 *) dpp - dp;
if (i != 0) {
vis_alignaddr((void *)(8 - offdst), 0);
c2 = vis_faligndata(dc02, dc01);
c1 = vis_faligndata(dc01, dc02);
}
else {
c1 = dc01;
c2 = dc02;
}
/* prepare the destination addresses */
spp = (mlib_d64 *) vis_alignaddr(sp, i);
sa1 = spp[0];
if (i != 0) {
/* generate edge mask for the start point */
emask = vis_edge8(dp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c2, sa);
vis_pst_8(da, dpp++, emask);
sa1 = sa2;
i += 8;
spp++;
}
#pragma pipeloop(0)
for (; i < amount - 16; i += 16) {
sa2 = spp[1];
ssa = vis_faligndata(sa1, sa2);
dpp[0] = VIS_CONSTLOGIC(c1, ssa);
sa3 = spp[2];
ssa1 = vis_faligndata(sa2, sa3);
dpp[1] = VIS_CONSTLOGIC(c2, ssa1);
sa1 = sa3;
dpp += 2;
spp += 2;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c1, sa);
vis_pst_8(da, dpp++, emask);
sa1 = sa2;
i += 8;
spp++;
}
if (i < amount) {
emask = vis_edge8(dpp, dend);
sa2 = spp[1];
sa = vis_faligndata(sa1, sa2);
da = VIS_CONSTLOGIC(c2, sa);
vis_pst_8(da, dpp++, emask);
}
}
sp = sl += stride;
dp = dl += strided;
}
}
}
return MLIB_SUCCESS;
}
/***************************************************************/
#ifdef __cplusplus
}
#endif /* __cplusplus */
#endif /* __MLIB_V_IMAGECONSTLOGIC_H */